Crop Vandalism by animals, namely wild boar, nilgai, monkeys and other wild animals. Animals visit the farms at time unknown. The unavailability of any patrolling, each yaer causes around 30-35% of crop losses to the farmers. High level fencing is too expensive and not durable. Treck Location: Grambharti, Ahmedabad, Gujrat.
We present an Neural Network Solution to the above problem. The software uses image recognisation and convulation networks to detects the things its capturing live from an external capturing, here we have used a web camera patrolling the field using a rover. The precision can further be improved by using a 360 degree view camera.
Apart from this we recommend a GPU based system(preferably NIVIDEA with CUDA installed) for smooth functioning of the project.
Follow the official documentation at: here
Dependencies : Python3, tensorflow 1.0, numpy, opencv 3.
Install Cython using pip install Cython (if not installed)
You can choose one of the following three ways to get started with darkflow.
-
Just build the Cython extensions in place. NOTE: If installing this way you will have to use
./flowin the cloned darkflow directory instead offlowas darkflow is not installed globally.python3 setup.py build_ext --inplace -
Let pip install darkflow globally in dev mode (still globally accessible, but changes to the code immediately take effect)
pip install -e . -
Install with pip globally
pip install . -
Update
Dependencies : Python3, opencv 3.
Follow the steps to get started with darknet.
-
Clone the repository using the following command.
git clone https://github.com/pjreddie/darknet -
cd darknet
-
make
-
Update
Testing the model.
-
Get a pre-trained model using the following command.
wget https://pjreddie.com/media/files/yolo.weights -
Run the detector.
./darknet detect cfg/yolo.cfg yolo.weights data/dog.jpg
Setting Up directories
Copy the darkflow installation into the darknet folder. Create a bin folder in the darknet and add the yolo.weights into it. You are good to go sweetheart!